PerceptionBench leaderboard
As of November 14, 2026, among the results listed here, Thinking Machines Inkling 2 Small has the highest reported PerceptionBench score, 64.7%, published by Thinking Machines, ahead of Qwen3.8 Max (0902) at 63.5%. 4 models have a PerceptionBench result in our records.
Scores as the model makers published them, not re-run by us; each row links to its announcement.
Best reported score
64.7%
T Thinking Machines Inkling 2 Small
Reported by Thinking Machines, November 9, 2026
4 models ranked · 3 makers
| Rank | Model | Score | Setting | Reported by | Date |
|---|---|---|---|---|---|
| 1 | T Inkling 2 Small | 64.7% | Not published | Thinking Machines | November 9, 2026 (model release) |
| 2 | Qwen3.8 Max (0902) | 63.5% | Not published | Qwen | August 3, 2026 |
| 3 | Kimi K3 | 58.5% | max | Moonshot AI | July 16, 2026 (model release) |
| 4 | Qwen3.7 Plus | 51.1% | Not published | Qwen | August 3, 2026 |
How much to trust these numbers
- Source
- Reported by the model makers themselves. Not re-run by LLMBenchmark or anyone independent.
- Settings
- Settings differ between makers. Check the Setting column before comparing two scores.
- Spread
- 64.7% to 51.1% across 4 models.
- Freshness
- Newest result published November 9, 2026; oldest July 16, 2026.
What PerceptionBench measures
PerceptionBench tests multimodal models on basic visual perception, with its score reflecting those capabilities; the search results do not specify how the score is calculated. Source: PerceptionBench: Evaluating Atomic Visual Perception